It’s time for another Fuzzy Friday today and this week I won’t discuss my previous ongoing series on 3PL but on another topic. Don’t worry, I’ll continue with Fixed Warehousing cost in rate cards in a couple of weeks. This week I would like to quickly discuss an example warehouse design approach through looking at different data elements and especially throughput.
Throughput Warehouse Design particularly looks at throughput volumes flowing through the different processes within the warehouse. Inbound containers are calculated into received pallets daily and in total. From these volumes we can flow put-away into different storage media across the DC and then replenishment & picking volumes are calculated, which give some indication for the warehouse solution.
However, throughput alone doesn’t tell as much about a suitable design. This is where order profiles come in to add more refinement to the design. I’ll try to explain this with the following examples:
Figure 1
Figure 1 gives total volumes per functional area (or task) within the warehouse but doesn’t tell us much about a potential solution. We can see pallets are received, put-away into rack and 3 separate pick faces; LSS Shelving for slow-movers, full case picks, pick-from-pallet for medium-moving full case picks and fast-moving full pallets being picked. However, it doesn’t tell us much about seasonality – a daily max gives some insight but when is that day and is it sustained? Typically, in 3PL when rate cards are populated, these total volumes are filled-in for commercial reasons, but it doesn’t say anything about an operationally viable solution.
In addition to looking at throughput summaries, it’s important to review order profile data and seasonality and using statistical methods to get a better understanding of the operation through more detailed data-analyses. For this we will use the average, average + 1 standard deviation and again maximum. (I’ll refer to the normal distribution or Bell curve for more information regarding these statistical evaluation methods)
Figure 2
Figure 3
Figure 4
Figure 2 & 3 give much more info with seasonality across the year. There is an obvious peak towards the end of the year which starts at September and ends at November. This is quite a typical profile for an FMCG B2B retailer who builds stock up-to September and then pushes this to stores for the holiday period.
Figure 4 give us more info on order profile and a potential solution. For medium-moving Full Case we have 29 orders producing 277 lines and almost 4,000 cases. That gives 14 cases / line and 136 cases / order which is very dense, every order produces almost 3 full pallets worth of material. A reach truck or triple-pallet ride-on with voice-picking for hands-free picking for all these full case orders would work well. As there is a lot of volume coming from the medium movers, an efficient full pallet replen is advised, hence pick-from-pallet is a suitable solution.
Slow-moving orders are smaller with less cartons per order and only 7 cases / line meaning less pick face touches to fulfill an order. There are many more orders and thus lines to fulfill daily which means a smaller pick face area to reduce travel time would work well. A denser pick face holding more SKUs in a smaller area, like levels of shelving over multiple bays would suit well for this operation.
Higher variance between average and maximum could have large implications for the design. Warehouse design should always have careful consideration of these aspects to allow for enough room and growth to flex up and down to facilitate volume.
Designing a warehouse purely on throughput volume and especially average volume is incredibly dangerous. Always look at how orders are built-up, how many lines are in each order and how the product is picked – full pallets, full layer, full case, or split case. Hourly surge (for example in E-comm) operations can have a large impact on warehouse design also. More data-analyses aspects contribute to good warehouse design, but I won’t be able to share them all this week. Hopefully you have gained some insight on data analytics methods and warehouse design.
Keep an eye on Fuzzy Friday for more insights and info.